Mountain Ok

Timothy Usher, Santa Fe Institute

[under construction]

The Mountain Ok family consists of as many as eight languages spoken in and to either side of New Guinea's central cordillera in Papua New Guinea's Sandaun and Western provinces and just west of the border with Indonesia's Papua province.

[under construction]

The internal classification of Mountain Ok is as follows:

Mountain Ok

Mian-Suganga

Mian

Suganga

Central Mountain Ok

Teléfól

Tifal

Kauwol

Faiwol

Setaman

Bimin

Urap (Urapmin) is not a distinct language but allied to the Tifal (Tifalmin) dialect of Tifal against the Atbalmin dialect (Healey 1964: 40-41.)

[under construction]

Champion (1926-1927) … Unkia (Faiwol)

Healey and Healey (1961) phonology of Teléfól

Healey (1962) kinship terms for Teléfól

Healey (1964) phonology of Teléfól

Healey (1964: 78-97) 120 comparative terms for Tifal of Atbalmin, Tifalmin and Wopkeimin villages, Kauwol, Faiwol of Imdalmin, Gipman, Faiwolmin, Angkiakmin, Wokfiakmin and Setaman villages, Bimin of Bimin and Kuskusmin villages, Teléfól, Mianmin (Mian) and Wagarabai (Suganga) and (pp. …) proto-Mountain Ok … (pp. 190-251) phonology of Teléfól

Healey (1964) Telefol quotatives

Healey (1965) Telefol clauses

Healey (1965) Telefol verb phrases

Healey (1965) Telefol syntax

Healey (1965) Telefol nominal phrases

Healey (1966) Telefol sentences

Healey (1970) reconstruction of proto-Awyu-Dumut includes (pp. …) some comparisons o proto-Mountain Ok

Healey (1974) classification of Telefol verbs

Healey and Healey (1977) dictionary of Teléfól

Healey (1981) medical terms for Telefol including body parts

Healey (1992) sketch phonology of Teléfól

Voorhoeve (2005) compares proto-Asmat-Kamoro and proto-Awyu-Dumut to proto-Mountain Ok

Steinkraus (1969) phonology of Tifal

Healey and Steinkraus (1972) dictionary and grammar notes for Tifal

Mecklenberg (1974) phonology of Faiwol (central dialect)

Mecklenberg (1992) sketch phonology of Faiwol

Boush and Boush (1974) phonology of Tifal

Boush (1975) grammar of Tifal

Boush (1975) verbal aspect in Tifal

Boush (1994) sketch phonology of Tifal

Smith and Weston (1974) phonology of Mianmin

Smith and Weston (1974) grammar of Mianmin

Smith (1977) Mianmin sentences

Weston (1977) Mianmin interrogatives

Smith and Weston (1986) dictionary of Mianmin

Weston (1992) sketch phonology of Mianmin

Weber (1992) sketch phonology of Bimin (unobtained)

Weber and Weber (1992) sketch phonology of Bimin (unobtained)

Weber (1997) grammar of Bimin (Bim dialect)

Weber (1997) dialect survey of Bimin

Weber and Whitney (1999) phonology of Bimin (Bim dialect)

Weber (2003) sketch phonology of Bimin

Fedden (2007) gender in Mian

Fedden (2007) grammar of Mian

Fedden (2011) grammar of Mian

Loughnane and Fedden (2011) comparison of Mountain Ok with Okspamin including protoforms

Van Cott and Spencer (2010) 190 comparative terms for Setaman of Seltamin, Ngisokabip and Baktamin villages, Faiwol of Selbang and Lowbip villages and Bimin of Dimtikin, Akalu, Sengen and Fagobiw villages

[under construction]

The reconstruction of Ok and of Mountain Ok in particular began with Healey's (1964) landmark dissertation. Much more data has been made available since then, allowing us to significantly improve and expand upon Healey's system.

Proto-Mountain Ok had 14 or 15 consonants … short vowels and … long vowels as follows:

In addition to the simple vowels given above, … diphthongs are found as follows:

Any consonant can occur initially. Voiced prenasalized apical stop /*nd/ occurs only initially, having elsewhere been merged with the reflexes of apical non-stop /*ɾ/. Rounded velarstops /*kʷ *ŋgʷ/ have been found only initially. Consonant clusters occur only accross morpheme boundaries in compunds.

Besides long vowels and diphthongs, vowels do not occur in sequence except across morpheme boundaries in compounds.

Only a restricted set of consonants occurs finally:

Mountain Ok roots are highly restricted in their phonotactic shapes. Only one long vowel can occur in any given root, more than one invariably indicating a compound. In disyllabic roots, the long vowel is overwhelmingly that of the second syllable, apparent exceptions likewise indicating compounds. Final vowels of either length occur freely in monosyllablic roots and infrequently in disyllables, but almost never on roots of three syllables or more, a fact which is historically intimately linked with the distribution of long vowels: it is nearly unavoidable that Mountain Ok descends from a language in which stress was penultimate and final vowels were common or mandatory, then these final vowels were destressed and dropped to yield a mostly consonant-final lexicon with final stress. The shapes of reconstructed uncompounded nominal roots include:

A small number of disyllabic nominals have a long vowel in the first syllable /*(C)VːCV(ː)C/. These are interpreted here as compounds, albeit in some instances highly lexicalized and possibly established as single roots by the time of proto-Mountain Ok.

Disyllabic roots with final short vowels are found only as verbs, where …

Two phonemic tones, high and low, interact with vowel length to yield four tones in some descendants. …

The most conspicuous difference between the system adopted here and that presented in Healey (1964) is the representation of the voiced stop series as prenasalized /*mb *nd *ŋg *ŋgʷ/ rather than Healey's /*b *d *g *gʷ/. There are four motivations for this treatment. First, these were their values in Proto-Ok, in Proto-Digul River-Ok and probably into primordial antiquity. Second, final labial /*mb/ has merged with /*m/, while final velar /*ŋg/ was realized as [*ŋ] and was likely still structurally equivalent initial /*ŋg/ [*g] at the time of Proto-Mountain Ok. The medial reflex of /*ŋg/ (Healey's /*ŋk/) is realized as [ŋg ŋ] in Bimin and Setaman and alternates between [ŋ g] in Faiwol (Mecklenburg 1974: 150,) seemingly mandating that at least /*ŋg/ was prenasalized. Lastly, according to Fedden (2007: 37,) Mian's initial voiced stops /b d g/ are slightly prenasalized, though less so than in Oksapmin; this is not attested in Telefol or Tifal (Healey and Healey 1961, Healey 1964, Steinkraus 1969, Boush and Boush 1974.)

The only challenge to equating velar nasal [ŋ], which almost only occurs finally, with /*ŋg/ is the existence of a rare and presumably innovative initial velar nasal /ŋ/ in several Mian words which is answered by /n/ in Telefol and Tifal; here we can assign the value /*ŋ/ with some confidence because /*ŋg/ is similarly merged with the reflex of /*nd/ [d] in the latter two languages. As only five examples of initial /*ŋ/ have been indentified so far, two species names and three conceivably onomatopoetic, none with known etymologies past Mountain Ok, its status is marginal at best, and may not date to proto-Mountain Ok.

The reinterpretation of Healey’s /*f/ as plain bilabial stop /*p/ likewise deserves an explanation. First, this was the underlying quality of its predecessor in proto-Ok. Second, apical and velar voicless stops are aspirated both initially and finally in Mianmin, Telefol, Tifal and Faiwol, while the final labial stop is realized as voicless [p] in Mian, sometimes in variation with [f], and as aspirated [pʰ] in Telefol, Tifal and Faiwol.(Healey 1964: 205, Boush and Boush 1974: 4-6, Mecklenberg 1974: CITE, 151-152, Smith and Weston 1974: 17-20.) Third, initial /f/ can be realized as affricate /pɸ/ in Telefol.(Healey 1964: 206.) The decisive factor is that what Healey and Healey, Boush and Boush and Smith and Weston interpret as the final allophone of /b/, and Mecklenberg as that of /w/, is historically (and arguably synchronically) the final allophone of what is initially and medially found as [f], the final reflexes of /*mb/ having merged with those of /*m/ by the time of proto-Mountain Ok. This allophony is nearly identical to that found in Oksapmin, in which Lawrence’s /p/ is realized as /ɸ/ initially, as /β/ intervocalically, and as /p/ finally (Lawrence CITE.)

Healey's /*c *cʷ/ are reducible to /*k *kʷ/ followed by /*i/.

Referring to the large number of reconstructed vowel correspondences, Healey (1964: 129) writes, “The apparent contrasts discussed above suggest a PMO [Proto-Mountain Ok] phonemic system much more complex than any of the daughter languages, and such a system is suspect because of this discrepancy.” Healey (1970) posited an equally or more complex system for proto-Awyu-Dumut (Central Digul River,) which likewise invites skepticism but, upon examination, not only holds up to scrutiny, but is shown to be nearly the only possible answer. Although most subfamilies of Ok have simplified this vowel inventory, these reconstructed values regularly correspond to mostly-isomorphic values in proto-Digul River and have thus been confirmed to a certainty beyond that which Healey (1964) was willing to grant them.

[under construction]

In the correpondence tables below, prevailing phonetic values for Mian are based upon Smith and Weston (1974, Weston 1992) and Fedden (2007,) for Telefol upon Healey and Healey (1961, Healey 1964,) for Tifal upon Steinkraus (1969) and Boush and Boush (1974, Boush 1994,) for Faiwol upon Mecklenberg (1974) and for Bimin upon Weber and Whitney (1999, Weber 2003.) These are correlated to Healey's (1964) proto-Mountain Ok values to support the comparison of the two systems.

Initial consonants correspond as follows:

Medial consonants correspond as follows:

Final consonants correspond as follows:

Consonant clusters in compounds correspond as follows:

…

Vowels and diphthongs correspond as follows:

These correspondences are exemplified as follows, with Mian attestations drawn from Smith and Weston (1974, 1986) Telefol from Healey and Healey (1977,) Tifal from Healey and Steinkraus (1972,) Faiwol from Mecklenberg (1974) and Bimin from Weber (1997, 2003, Weber and Whitney 1999) unless otherwise specified. …

Initial consonants … :

Initial bilabial nasal /*m/ is retained as such in all descendants:

Initial apical nasal /*n/ is retained as such in all descendants:

Initial velar nasal /*ŋ/ …:

Initial bilabial voiceless stop /*p/ … :

… /*p/ … In accordance with these rules, Smith and Weston (1986: 16) give only one word in which unoccluded /h/ is followed by high front vowel /i/ and one where it is followed by high back rounded /u/; conversely (p. 12) only two native words are given in which initial labiodental /f/ is followed by mid front vowel /e/. … :

Initial apical voiceless stop /*t/ is retained as such in all descendants:

Initial voiceless laminal /*s/ is retained as such in all descendants:

Initial velar voiceless stop /*k/ …:

When followed by high front vowel /*i *iː/, velar voiceless stop /*k/ …:

Due to this merger, a number of roots which are attested only in Telefol and Tifal are indeterminate between initial apical /*t/ and initial velar /*k/:

There is one very salient exception to this rule …:

… /*k/ …:

Initial rounded velar voiceless stop /*kʷ/ …:

When followed by high front vowel /*i *iː/, rounded velar voiceless stop /*kʷ/ …:

Initial bilabial voiced prenasalized stop /*mb/ …:

In a few examples, bilabial voiced prenasalized /*mb/ … /*p/ …:

Initial apical voiced prenasalized stop /*nd/ …:

Initial velar voiced prenasalized stop /*ŋg/ …:

Due to this merger, a number of roots which are attested only in Telefol and Tifal are indeterminate between initial apical /*nd/ and initial velar /*ŋg/:

Initial rounded velar voiced prenasalied stop /*ŋgʷ/ …:

Initial bilabial non-stop /*w/ …:

Initial apical non-stop /*ɾ/ …:

Initial palatal non-stop /*j/ …:

Medial consonants …:

Medial bilabial nasal /*m/ is retained as such in all descendants:

Medial apical nasal /*n/ is generally retained as such in all descendants:

… /*n/ … dissimilated to bilabial nasal /m/ when …:

Medial bilabial voiceless stop /*p/ …:

Medial apical voiceless stop /*t/ is retained as such in all descendants:

Medial voiceless laminal /*s/ is retained as such in all descendants:

Medial velar voiceless stop /*k/ …:

… /*k/ …:

Medial bilabial voiced prenasalized stop /*mb/ …:

Apical voiced prenasalized stop /*nd/ is not found root-medially because Proto-Ok medial /*nd/ has been merged with the reflexes of apical non-stop /*ɾ/ (below.)

Medial velar voiced prenasalized stop /*ŋg/ …:

Due to this merger … /*ŋg/ …:

Medial bilabial non-stop /*w/ …:

Medial apical non-stop /*ɾ/ …:

In a few examples …:

Medial palatal non-stop /*j/ …:

Only a restricted set of consonants occurs finally due to mergers of Proto-Ok consonants in this position. These are nasals /*m *n/, voiceless /*p *t *s *k/ and voiced prenasalized /*ŋg/.

Final bilabial nasal /*m/ is retained as such in all descendants:

Final apical nasal /*n/ is generally retained as such in all descendants:

One apparent exception …:

Final bilabial voiceless stop /*p/ …:

Final apical voiceless stop /*t/ …:

Final voiceless laminal /*s/ …:

When root-final voiceless laminal /*s/ is the first member of a compound …:

Final velar voiceless stop /*k/ …:

… /*k/ …:

There is no final bilabial voiced prenasalized stop /*mb/ because Proto-Ok final /*mb/ has merged with nasal /*m/. Some compounds, presumably dating to before the sounds were merged, reflect final /*m/ as /*mb/:

There is no final apical voiced prenasalized stop /*nd/ because Proto-Ok /*nd/ has merged with non-stop /*ɾ/.

Final velar voiced prenasalized stop /*ŋg/ …:

… final /*w *j/ …

… clusters in compounds …

…

Vowels and diphthongs …

Short high front vowel /*i/ …:

One example …:

When short high front /*i/ is preceded by initial bilabial non-stop /*w/ …:

Long high front vowel /*iː/ …:

Short high central rounded vowel /*ʉ/ …:

Long high central rounded vowel /*ʉː/ …:

Short high back rounded vowel /*u/ …:

When short high back rounded /*u/ …:

Long high back rounded /*uː/ …:

Long high mid front vowel /*e̞ː/ …:

When long high front /*e̞ː/ occurs word-initially …:

…:

Short high mid back rounded vowel /*o̞/ …:

In one example, … /*o̞/ …:

Long high mid back rounded vowel /*o̞ː/ …:

Short low mid front vowel /*ɛ/ …:

… short low mid front /*ɛ/ … /*w/ …:

… short low mid front /*ɛ/ … /*p/ … [w] …:

… /*ɛ/ …:

Evidence for a long counterpart to low mid front /ɛ/ is very scant …:

Short low mid back rounded vowel /*ɔ/ … nearly identical to the reflexes of short low mid front /*ɛ/ excepting Mian …:

When short low mid rounded /*ɔ/ is preceded by initial apical non-stop /*ɾ/ …:

When short low mid rounded /*ɔ/ is preceded by initial /*i/ in a previous syllable …:

Several of the terms above are historically compounds with /*ɔːk/ “water” in which long /*ɔː/ has been reduced to /*ɔ/.

Long low mid back rounded vowel /*ɔː/ …:

When long low mid rounded /*ɔː/ is preceded by initial /*i/ in a previous syllable …:

Short low central vowel /*a/ …:

When short low central /*a/ is preceded by initial bilabial non-stop /*w/ …:

… short low central /*a/ … /*w/ …:

… short low central /*a/ … /*p/ … [w] …:

Long low central vowel /*aː/ …:

… diphthongs ….

… /*eⁱ/ …:

… /*oⁱ/ …:

… /*aⁱ/ … Unlike /*eⁱ/ (above,) following /*aⁱ/ induces the loss of initial velar voiceless stop /*k/ in Mian:

Due to these mergers, a number of roots which are attested only in Mian, Telefol and/or Tifal are indeterminate between /*eⁱ/ and /*aⁱ/:

… /*eᵘ/ …:

… /*oᵘ/ …:

… /*aᵘ/ …:

… /*aᵘ/ …:

… /*aᵉ/ …:

… /*aᵉ/ …:

… /*aᵒ/ …:

… destressed vowels … /*i *a/ ….

When short high front /*i/ is destressed and followed in the next syllable by a stressed vowel other than /*i *iː/ …:

In one example, …:

When short high front /*i/ …:

When short low central /*a/ is destressed and followed in the next syllable by a stressed vowel other than /*a *aː/ …:

When destressed short low central /*a/ is preceded by initial rounded velars /*kʷ *ŋgʷ/ …

When followed by bilabial non-stop /*w/, destressed /*a/ …:

When followed by palatal non-stop /*j/, destressed /*a/ …:

When initial short low central /*a/ is followed by …:

… word shape :

…:

… tone ….

…: